Ergonomically Intelligent Physical Human-Robot Interaction: Postural Estimation, Assessment, and Optimization

Abstract

Ergonomics and human comfort are essential concerns in physical human-robot interaction applications, and common practical methods either fail in estimating the correct posture due to occlusion or suffer from less accurate ergonomics models in their postural optimization methods. Instead, we propose a novel framework for posture estimation, assessment, and optimization for ergonomically intelligent physical human-robot interaction. We show that we can estimate human posture solely from the trajectory of the interacting robot. We propose DULA, a differentiable ergonomics model, and use it in gradient-free postural optimization for physical human-robot interaction tasks such as co-manipulation and teleoperation. We evaluate our framework through human and simulation experiments.

NeuroErgo: A Deep Neural Network Method to Improve Postural Optimization for Ergonomic Human-Robot Collaboration

“…Although the essence of their idea is the same as the current paper, their method handles straightforward scenarios and is not suitable for more complex cases in real applications. Second, a pre-printed paper [24] that attempted to apply neural networks to obtain a differentiable RULA and postponed REBA modeling as future work. Although the essence of their work is quite similar to NeuroErgo, they did not discuss the corresponding neural network's hyperparameters selection, training, testing, data generation, and potential problems for ReLU activation function as it is not differentiable at point zero.…”

Occlusion-Robust Multi-Sensory Posture Estimation in Physical Human-Robot Interaction

“…To improve ergonomics and lower the risk of WMSDs in teleoperation, we introduced a ergonomically intelligent teleoperation System in [70,73], in which the teleoperation system intelligently estimates the human posture during interaction with the leader robot (more details in [71]), assesses the posture using ergonomics risk assessment tools, and if the risk of WMSDs is high, provides an ergonomically optimal postural correction for the human to complete the task (more details in [72]). We also introduced different types of autonomous postural correction in teleoperation using the leader robot.…”